Model comparison
GLM-5.3 vs Inkling-Small
GLM-5.3 is the stronger model overall, scoring 54.6 to 46.5 on the Noometry Index. Inkling-Small costs 3.4× less per token, which makes it the better buy when GLM-5.3's lead doesn't matter for your workload.
Last verified . 30 shared benchmarks.
Summary
- They share 30 benchmarks with published results for both. GLM-5.3 scores higher in 8 categories and Inkling-Small in 0 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where GLM-5.3 leads 62.1 to 44.9.
- The biggest single-benchmark swing is ProofBench: 49% for GLM-5.3 and 6% for Inkling-Small.
- Inkling-Small is cheaper at $0.45 / $1.20 per million input/output tokens, against $1.40 / $4.40 for GLM-5.3.
- GLM-5.3 accepts more context: 1M tokens versus 524K.
Side by side
| GLM-5.3 | Inkling-Small | |
|---|---|---|
| Provider | Z.ai (Zhipu) | Thinking Machines Lab |
| Noometry Index | 54.6 | 46.5 |
| Released | 2026-08-14 | 2026-07-15 |
| Weights | Open | Open |
| Context window | 1M | 524K |
| Max output | 131K | 1.05M |
| Input $ / M tokens | $1.40 | $0.45 |
| Output $ / M tokens | $4.40 | $1.20 |
| Results tracked | 41 | 33 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GLM-5.3 leads
GLM-5.3: 59.0 (#12), Inkling-Small: 43.3 (#80)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| LMArena WebDev | 1623 | 1408 |
| SciCode | 59% | 48.7% |
| LMArena Coding | 1499 | 1451 |
| DeepSWE | 69% | — |
| FrontierCode | 40.1% | — |
| FrontierSWE | 30.2% | — |
| WeirdML | 75.4% | — |
| ALE-Bench | 1,317 | — |
Agentic & Tool Use Not comparable
GLM-5.3: 36.3 (#34), Inkling-Small: —
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| APEX-Agents | 56.6% | — |
| Vending-Bench 2 | 8,164 | — |
Reasoning GLM-5.3 leads
GLM-5.3: 45.2 (#42), Inkling-Small: 39.0 (#53)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| CritPt | 19.1% | 8.3% |
| Chess Puzzles | 21% | 18% |
| LMArena Hard Prompts | 1491 | 1424 |
| Mystery Game Puzzles | 33% | 6% |
| Epoch Capabilities Index | 155.56 | 150.13 |
| ARC-AGI-2 | — | 40.1% |
| NYT Connections (extended) | 74.2% | — |
| ARC-AGI-1 | — | 84% |
| DTBench | 87.7% | — |
| LMCA | 55.5% | — |
| Bench to the Future 3 | 0.15 | — |
Math GLM-5.3 leads
GLM-5.3: 62.1 (#27), Inkling-Small: 44.9 (#65)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| FrontierMath (Tiers 1-3) | 68.8% | 46.3% |
| FrontierMath Tier 4 | 29.3% | 17.1% |
| OTIS Mock AIME 2024-2025 | 91.1% | 90% |
| ProofBench | 49% | 6% |
| LMArena Math | 1493 | 1453 |
Knowledge GLM-5.3 leads
GLM-5.3: 58.3 (#34), Inkling-Small: 48.3 (#71)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| GPQA Diamond | 90.9% | 88.5% |
| SimpleQA Verified | 41% | 19.1% |
| LMArena Expert | 1512 | 1442 |
Multimodal Not comparable
GLM-5.3: —, Inkling-Small: 38.4 (#59)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| LMArena Vision | — | 1235 |
Multilingual GLM-5.3 leads
GLM-5.3: 55.9 (#27), Inkling-Small: 51.7 (#102)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| LMArena Non-English | 1457 | 1400 |
| LMArena Chinese | 1523 | 1458 |
| LMArena French | 1499 | 1438 |
| LMArena German | 1505 | 1412 |
| LMArena Japanese | 1438 | 1399 |
| LMArena Korean | 1478 | 1355 |
| LMArena Russian | 1460 | 1392 |
| LMArena Spanish | 1444 | 1426 |
Instruction Following GLM-5.3 leads
GLM-5.3: 77.8 (#15), Inkling-Small: 73.9 (#112)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| LMArena Instruction Following | 1481 | 1397 |
Long Context GLM-5.3 leads
GLM-5.3: 45.5 (#32), Inkling-Small: 42.7 (#113)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| LMArena Longer Query | 1485 | 1401 |
Writing & Preference GLM-5.3 leads
GLM-5.3: 75.9 (#6), Inkling-Small: 59.8 (#102)
| Benchmark | GLM-5.3 | Inkling-Small |
|---|---|---|
| LMArena Text | 1473 | 1413 |
| LMArena Creative Writing | 1459 | 1330 |
| EQ-Bench Creative Writing | 2075 | 1491 |
| LMArena Multi-Turn | 1473 | 1420 |
Frequently asked questions
Is GLM-5.3 better than Inkling-Small?
GLM-5.3 is the stronger model overall, scoring 54.6 to 46.5 on the Noometry Index. Inkling-Small costs 3.4× less per token, which makes it the better buy when GLM-5.3's lead doesn't matter for your workload.
Which is cheaper, GLM-5.3 or Inkling-Small?
Inkling-Small is cheaper. It lists at $0.45 per million input tokens and $1.20 per million output tokens; GLM-5.3 lists at $1.40 and $4.40.
Is GLM-5.3 or Inkling-Small better for coding?
GLM-5.3 scores higher on coding benchmarks: 59.0 versus 43.3 in the Noometry coding category.
Which has the bigger context window?
GLM-5.3 does, with 1M tokens against 524K.
How many benchmarks do GLM-5.3 and Inkling-Small share?
30 benchmarks have published results for both models. GLM-5.3 has 41 scored results on Noometry and Inkling-Small has 33.